rtdetr-v2-r50-finetune-2

This model is a fine-tuned version of PekingU/rtdetr_v2_r50vd on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 8.2861
Map: 0.5775
Map 50: 0.9118
Map 75: 0.6739
Map Small: 0.539
Map Medium: 0.6557
Map Large: -1.0
Mar 1: 0.335
Mar 10: 0.6696
Mar 100: 0.6706
Mar Small: 0.639
Mar Medium: 0.7406
Mar Large: -1.0

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 300
num_epochs: 80

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large
No log	1.0	105	23.3284	0.069	0.165	0.0459	0.0434	0.1163	-1.0	0.1333	0.3729	0.486	0.3726	0.6219	-1.0
No log	2.0	210	10.6884	0.4152	0.7631	0.3744	0.3166	0.535	-1.0	0.343	0.61	0.6417	0.5663	0.7322	-1.0
No log	3.0	315	9.0848	0.4805	0.8412	0.5038	0.3894	0.5907	-1.0	0.3645	0.614	0.6558	0.5869	0.7384	-1.0
No log	4.0	420	8.5359	0.4691	0.8554	0.4428	0.3957	0.5599	-1.0	0.3685	0.5991	0.629	0.5851	0.6815	-1.0
43.2256	5.0	525	8.3908	0.492	0.8614	0.5158	0.4063	0.5947	-1.0	0.376	0.6053	0.6293	0.5549	0.7185	-1.0
43.2256	6.0	630	8.3865	0.4815	0.8604	0.4839	0.3984	0.582	-1.0	0.3773	0.6178	0.6302	0.5749	0.6966	-1.0
43.2256	7.0	735	8.6199	0.482	0.8483	0.4817	0.3954	0.5907	-1.0	0.3738	0.6128	0.6374	0.5754	0.7116	-1.0
43.2256	8.0	840	8.2561	0.4778	0.8792	0.4492	0.3959	0.5866	-1.0	0.3798	0.6031	0.6224	0.5657	0.6904	-1.0
43.2256	9.0	945	8.4404	0.4904	0.8793	0.5023	0.3983	0.5978	-1.0	0.3729	0.615	0.6343	0.572	0.7089	-1.0
11.2853	10.0	1050	9.0564	0.4823	0.8876	0.4404	0.391	0.597	-1.0	0.366	0.5938	0.6097	0.5411	0.6918	-1.0
11.2853	11.0	1155	8.8622	0.465	0.8691	0.4306	0.3762	0.5931	-1.0	0.3592	0.6081	0.6112	0.548	0.687	-1.0
11.2853	12.0	1260	8.8839	0.4785	0.8801	0.4591	0.3984	0.5883	-1.0	0.3717	0.6146	0.6206	0.5457	0.7103	-1.0
11.2853	13.0	1365	8.7132	0.4835	0.8776	0.4713	0.4058	0.5769	-1.0	0.371	0.5935	0.5963	0.5274	0.6788	-1.0
11.2853	14.0	1470	8.8340	0.4892	0.8748	0.499	0.4005	0.5965	-1.0	0.372	0.6106	0.6227	0.556	0.7027	-1.0
9.4718	15.0	1575	9.2052	0.478	0.8641	0.4606	0.3822	0.5908	-1.0	0.3648	0.5931	0.5953	0.5006	0.7089	-1.0
9.4718	16.0	1680	9.3522	0.4933	0.8737	0.49	0.393	0.6028	-1.0	0.3779	0.6006	0.6019	0.5291	0.689	-1.0
9.4718	17.0	1785	9.1062	0.4855	0.8772	0.4926	0.3918	0.5949	-1.0	0.3766	0.5919	0.5931	0.5189	0.6822	-1.0
9.4718	18.0	1890	9.0254	0.4931	0.8781	0.4905	0.4071	0.5928	-1.0	0.3748	0.591	0.5966	0.5211	0.687	-1.0
9.4718	19.0	1995	9.1593	0.4935	0.8884	0.4932	0.4078	0.5946	-1.0	0.3788	0.6009	0.6031	0.5291	0.6918	-1.0
8.281	20.0	2100	9.4549	0.4828	0.8572	0.4841	0.3928	0.5883	-1.0	0.3729	0.5857	0.5903	0.508	0.689	-1.0
8.281	21.0	2205	9.4448	0.4886	0.8637	0.4955	0.4007	0.594	-1.0	0.3692	0.5956	0.5966	0.5126	0.6973	-1.0
8.281	22.0	2310	9.2595	0.4848	0.8669	0.4756	0.3986	0.5922	-1.0	0.3592	0.5854	0.5872	0.4954	0.6973	-1.0
8.281	23.0	2415	9.3288	0.4997	0.8727	0.5155	0.4094	0.6037	-1.0	0.3766	0.5922	0.5931	0.5091	0.6938	-1.0
7.4164	24.0	2520	9.0350	0.4977	0.8798	0.4747	0.4099	0.5973	-1.0	0.3776	0.5947	0.595	0.5274	0.676	-1.0
7.4164	25.0	2625	10.2109	0.4888	0.8758	0.477	0.3975	0.5914	-1.0	0.3651	0.59	0.5916	0.504	0.6966	-1.0
7.4164	26.0	2730	9.9737	0.4891	0.8838	0.4853	0.4142	0.5772	-1.0	0.3788	0.595	0.5956	0.5291	0.6753	-1.0
7.4164	27.0	2835	9.5459	0.4942	0.8696	0.5067	0.4089	0.5979	-1.0	0.3757	0.5925	0.5944	0.5229	0.6801	-1.0
7.4164	28.0	2940	9.4796	0.492	0.8777	0.4905	0.4062	0.59	-1.0	0.3741	0.5863	0.5879	0.5029	0.6897	-1.0
6.774	29.0	3045	9.4067	0.4875	0.8707	0.4821	0.4021	0.589	-1.0	0.3788	0.5879	0.5894	0.5063	0.689	-1.0
6.774	30.0	3150	10.6548	0.4821	0.8661	0.4755	0.3883	0.5892	-1.0	0.3673	0.5872	0.5875	0.496	0.6973	-1.0
6.774	31.0	3255	9.9579	0.4911	0.8802	0.4725	0.4014	0.5923	-1.0	0.3713	0.5822	0.5835	0.4943	0.6904	-1.0
6.774	32.0	3360	10.0836	0.4882	0.8661	0.4812	0.3985	0.59	-1.0	0.3707	0.5888	0.5897	0.512	0.6829	-1.0
6.774	33.0	3465	9.8357	0.4942	0.8775	0.484	0.4129	0.593	-1.0	0.3713	0.5838	0.5847	0.4954	0.6918	-1.0
6.2631	34.0	3570	9.7274	0.5004	0.8814	0.5101	0.4084	0.6029	-1.0	0.3748	0.591	0.5928	0.508	0.6945	-1.0
6.2631	35.0	3675	9.6835	0.494	0.8814	0.4834	0.4053	0.5938	-1.0	0.371	0.5882	0.5903	0.5063	0.6911	-1.0
6.2631	36.0	3780	9.7898	0.4949	0.8873	0.4624	0.4078	0.5918	-1.0	0.3773	0.5894	0.59	0.504	0.6932	-1.0
6.2631	37.0	3885	9.9264	0.4918	0.8852	0.4671	0.407	0.5891	-1.0	0.3741	0.595	0.5953	0.5131	0.6938	-1.0
6.2631	38.0	3990	10.3368	0.4959	0.8764	0.4884	0.404	0.5994	-1.0	0.3685	0.5928	0.5944	0.5114	0.6938	-1.0
5.9018	39.0	4095	10.2500	0.491	0.8831	0.4886	0.4037	0.5897	-1.0	0.3713	0.5863	0.5885	0.4994	0.6952	-1.0
5.9018	40.0	4200	9.9967	0.4981	0.8757	0.4928	0.4127	0.5993	-1.0	0.3785	0.5875	0.5875	0.4954	0.6979	-1.0
5.9018	41.0	4305	10.4797	0.4975	0.8756	0.4861	0.4107	0.6001	-1.0	0.3732	0.5938	0.595	0.508	0.6993	-1.0
5.9018	42.0	4410	9.9769	0.4994	0.8745	0.5072	0.4156	0.6038	-1.0	0.3707	0.5972	0.5978	0.5103	0.7027	-1.0
5.5449	43.0	4515	10.1072	0.4981	0.8754	0.4884	0.4089	0.5997	-1.0	0.3745	0.5879	0.5885	0.5006	0.6938	-1.0
5.5449	44.0	4620	10.6945	0.4903	0.8774	0.4766	0.399	0.5913	-1.0	0.3645	0.586	0.5866	0.5006	0.6897	-1.0
5.5449	45.0	4725	10.6691	0.5012	0.8824	0.4983	0.4104	0.6008	-1.0	0.3735	0.5947	0.595	0.5086	0.6986	-1.0
5.5449	46.0	4830	10.4118	0.4989	0.8846	0.4881	0.414	0.5973	-1.0	0.3723	0.5991	0.6003	0.5177	0.6993	-1.0
5.5449	47.0	4935	10.4624	0.4991	0.8739	0.4881	0.4146	0.5956	-1.0	0.3729	0.5963	0.5963	0.5211	0.6863	-1.0
5.206	48.0	5040	9.8917	0.4872	0.876	0.4655	0.3984	0.5895	-1.0	0.3717	0.5935	0.5935	0.516	0.6863	-1.0
5.206	49.0	5145	10.0398	0.4948	0.8749	0.4885	0.4029	0.5992	-1.0	0.3785	0.5885	0.5903	0.5063	0.6911	-1.0
5.206	50.0	5250	10.7227	0.4896	0.8664	0.4685	0.3891	0.6	-1.0	0.3654	0.5907	0.5913	0.5063	0.6932	-1.0
5.206	51.0	5355	10.1973	0.5004	0.8669	0.4891	0.4054	0.6053	-1.0	0.3766	0.5907	0.5919	0.5074	0.6932	-1.0
5.206	52.0	5460	11.0716	0.4932	0.8663	0.4763	0.4017	0.5957	-1.0	0.3692	0.5928	0.5928	0.5114	0.6904	-1.0
4.9513	53.0	5565	10.3798	0.502	0.871	0.4935	0.4062	0.6096	-1.0	0.3723	0.5913	0.5919	0.5029	0.6986	-1.0
4.9513	54.0	5670	10.9243	0.4987	0.8694	0.493	0.4074	0.6011	-1.0	0.376	0.5928	0.5928	0.5074	0.6952	-1.0
4.9513	55.0	5775	10.5659	0.4945	0.8774	0.4761	0.3974	0.6031	-1.0	0.3713	0.591	0.5925	0.5074	0.6945	-1.0
4.9513	56.0	5880	10.9075	0.4899	0.8656	0.483	0.3939	0.6004	-1.0	0.3654	0.591	0.5931	0.5091	0.6938	-1.0
4.9513	57.0	5985	11.2689	0.4951	0.8688	0.4633	0.3968	0.6013	-1.0	0.3695	0.5916	0.5931	0.5103	0.6925	-1.0
4.6313	58.0	6090	11.1352	0.5012	0.875	0.4975	0.4062	0.6071	-1.0	0.3735	0.5938	0.5941	0.5103	0.6945	-1.0
4.6313	59.0	6195	11.1320	0.4981	0.8713	0.4886	0.4048	0.6037	-1.0	0.372	0.5919	0.5919	0.5051	0.6959	-1.0
4.6313	60.0	6300	11.0104	0.5	0.873	0.4906	0.4063	0.6049	-1.0	0.3769	0.5925	0.5928	0.508	0.6945	-1.0
4.6313	61.0	6405	11.1602	0.4988	0.866	0.4853	0.4034	0.6079	-1.0	0.3773	0.5907	0.5919	0.504	0.6973	-1.0
4.4386	62.0	6510	11.1256	0.4983	0.8739	0.4736	0.4043	0.6023	-1.0	0.3741	0.59	0.5922	0.5086	0.6925	-1.0
4.4386	63.0	6615	11.1695	0.501	0.8734	0.4849	0.4021	0.6079	-1.0	0.3754	0.5944	0.5972	0.5109	0.7007	-1.0
4.4386	64.0	6720	11.1439	0.4997	0.8694	0.4972	0.4033	0.6063	-1.0	0.3779	0.5938	0.595	0.5097	0.6973	-1.0
4.4386	65.0	6825	11.4394	0.5	0.8654	0.4998	0.3992	0.6087	-1.0	0.376	0.5966	0.5981	0.5143	0.6986	-1.0
4.4386	66.0	6930	11.5553	0.4975	0.8661	0.4879	0.3968	0.6057	-1.0	0.371	0.595	0.595	0.5109	0.6959	-1.0
4.1871	67.0	7035	11.5702	0.4963	0.8707	0.4764	0.3954	0.6032	-1.0	0.3717	0.5947	0.5947	0.5109	0.6952	-1.0
4.1871	68.0	7140	11.5148	0.4968	0.868	0.4888	0.3942	0.6061	-1.0	0.3704	0.5944	0.5944	0.5086	0.6973	-1.0
4.1871	69.0	7245	11.5774	0.4961	0.8688	0.4725	0.3971	0.6021	-1.0	0.3713	0.5931	0.5944	0.512	0.6932	-1.0
4.1871	70.0	7350	11.6360	0.5004	0.87	0.4746	0.3959	0.6095	-1.0	0.3757	0.5941	0.595	0.5097	0.6973	-1.0
4.1871	71.0	7455	11.6617	0.4989	0.8667	0.4888	0.402	0.6036	-1.0	0.3735	0.5922	0.5931	0.5069	0.6966	-1.0
4.0014	72.0	7560	11.7640	0.4972	0.8644	0.4885	0.3962	0.6047	-1.0	0.3704	0.5935	0.5935	0.5074	0.6966	-1.0
4.0014	73.0	7665	11.9004	0.4981	0.8678	0.4824	0.3949	0.6049	-1.0	0.3707	0.5953	0.5963	0.5137	0.6952	-1.0
4.0014	74.0	7770	11.9330	0.4964	0.8637	0.4995	0.3946	0.6065	-1.0	0.3704	0.595	0.595	0.5074	0.7	-1.0
4.0014	75.0	7875	11.8106	0.4954	0.863	0.4861	0.3927	0.6071	-1.0	0.371	0.5931	0.5935	0.5063	0.6979	-1.0
4.0014	76.0	7980	12.0478	0.4931	0.8663	0.4775	0.393	0.6004	-1.0	0.3695	0.5897	0.5907	0.5063	0.6918	-1.0
3.7902	77.0	8085	11.8309	0.4947	0.8677	0.4817	0.3949	0.6028	-1.0	0.3732	0.5916	0.5919	0.5069	0.6938	-1.0
3.7902	78.0	8190	11.9550	0.4952	0.8688	0.4891	0.3949	0.6035	-1.0	0.3692	0.59	0.5919	0.5069	0.6938	-1.0
3.7902	79.0	8295	11.9819	0.4964	0.8685	0.4801	0.3962	0.6051	-1.0	0.3723	0.591	0.5935	0.5086	0.6952	-1.0
3.7902	80.0	8400	12.0452	0.4941	0.8673	0.48	0.3941	0.6025	-1.0	0.3704	0.5916	0.5925	0.5074	0.6945	-1.0

Framework versions

Transformers 4.56.1
Pytorch 2.8.0+cu128
Datasets 4.0.0
Tokenizers 0.22.0

dariacuna/rtdetr-v2-r50-finetune-2

rtdetr-v2-r50-finetune-2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种